I'm back at work from vacation life and I've actually missed it. Having hours to myself and working at technical problems that I mostly enjoy. I won't claim that fiddling with environment variables and their propagation through Docker Compose, Docker and a Phoenix release was completely filling me with joy. But I've been satisfied to be able to sit down and do this stuff again.

Another stream on the YouTube channel today. Due to logistics it is at 13.30 CEST. I haven't decided if I should keep hacking at some of the final pieces for the Photo thing or if we are going to be poking at some more experimental fun stuff. Considering poking at collaborative editing with OT.

A new blog post is out: It is not about Elixir

A new episode of Regular Programming is out, about onboarding. Creeping up towards double digit episode count. If you listen, do let me know how you like it.

This could get really annoying

A few newsletters back I touched on wanting to build more of my own software for publishing and such. There's more things I want to break the mold on and if I can even pull it off, which should be possible, it will be a very inconvenient transition for me. The end goal should put me square in the middle of lining up my business with my values though.

So I want to replace most of the day-to-day software I use with open software or small-scale trustworthy proprietary software. I'm already using open source to a very large extent. But there are parts of my business stack that are very big platform. It doesn't take a lot to guess it. E-mail and calendar are my biggest ones. Documents is another. So I use G-Suite (Google Apps) for my business domain. It works very well.

The change I want to make is not primarily for any practical reason. It is value-driven and based on wanting to escape a gravity well that will only pull harder with the weight of my usage. I don't like these large companies. I distrust them and would rather support some other effort.

There are places where I don't plan changes, where the entire point of the usage is to interface with proprietary or specialized systems. Stripe, there's no open path to credit cards. They aren't open. Bookkeeping, I'll roll with whatever my accountant likes that doesn't completely suck.

There is no reason why my email has to be in a proprietary system. There is no reason why my calendar has to. Or my documents.

I don't want to build these things myself. And I won't be making an instant change. This is the direction I want to go. And I want good solutions that solve my entire use-case and won't unduly piss of the people I work with. I'll still have a Google account, maybe I'll have to keep my business domain as legacy. We'll see. But I want to see if I can do things the way I want them done, rather than the convenient way. That's part of my value system. There's just a lot of activation energy needed to roll this particular ball.

So what am I looking to replace and get rid of?

• Gmail - Basic email needs.
• Google Calendar - Normal calendar needs. Invites in both directions, so probably mail integration?
• Google Drive - File storage, archive and reference
• Google Docs - Collaborative editing of text documents, might actually be mostly able to run this off of HackMD or similar. Not as important for spreadsheets, etc. Might not be able to abandon entirely for interfacing with clients.
•  Slack - I don't think I want it for my own business, I'll never be rid of it for client work and other things.
• Chromium - Might want to replace it, might be more feasible when I'm not doing so much Google.
• YouTube - Only partially. So YouTube is a way for people to discover what I do. I don't think I should abandon that as an entryway. But I'm looking at some interesting options.

I think the big two right now are email and calendar as I use those heavily and the calendar runs my day. Replacement clients for that really need to be quite good. My calendar notifications are critical infrastructure.

I think Drive/Docs is a sleeper challenge. I find Drive super-clunky but it does work really well.

Let me know whatever suggestions you have. Self-hosting is an option, paid versions of open solutions, also a good option. Or am I entirely unhinged to put effort in this direction? I don't believe I am but the return is not in money.

You can reach me at lars@underjord.io or Twitter where I'm @lawik.

Thank you for reading, I appreciate your attention.

- Lars Wikman